On Friday, 21 February 2020 08:53:33 PST Oswald Buddenhagen wrote:
> On Fri, Feb 21, 2020 at 08:29:54AM -0800, Thiago Macieira wrote:
> >can we deprecate setupChildProcess() in Qt 5.15 and *remove it* in 6.0?
>
> at least, we should. ;)
> but let's bikeshed the name of the callback setter, which needs to be
> added at the same time -
> setChildProcess{Modifier,Initializer,Setup,...}() ^^
I even thought of making it a signal, like QCoreApplication::aboutToQuit:
QProcess::childAboutToStart(). That allows for N hooks to be considered, not
just one, and derived classes like KProcess could insert their own hooks
independent of their users.
But that is probably a bad idea, as the machinery for emitting signals isn't
designed to be async-signal-safe. For any receiver in another thread, there's
also a mutex, which we can't assume we can lock.
If a derived class needs its own child code, it could override the setter and
wrap calling the user's inside its own.
--
Thiago Macieira - thiago.macieira (AT) intel.com
Software Architect - Intel System Software Products
_______________________________________________
Development mailing list
Development@qt-project.org
https://lists.qt-project.org/listinfo/development
